pfinet: Add x86_64 checksum support

2 files changed, 278 insertions, 0 deletions

diff --git a/pfinet/linux-src/arch/x86_64/lib/checksum.c b/pfinet/linux-src/arch/x86_64/lib/checksum.c
new file mode 100644
index 00000000..01f28eef
--- /dev/null
+++ b/pfinet/linux-src/arch/x86_64/lib/checksum.c
@@ -0,0 +1,169 @@
+/*
+ * arch/x86/lib/checksum.c
+ *
+ * This file contains network checksum routines that are better done
+ * in an architecture-specific manner due to speed..
+ */
+ 
+#include <linux/string.h>
+
+#include <asm/byteorder.h>
+
+static inline unsigned short from64to16(unsigned long x)
+{
+	/* add up 32-bit words for 33 bits */
+	x = (x & 0xffffffff) + (x >> 32);
+	/* add up 16-bit and 17-bit words for 17+c bits */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up 16-bit and 2-bit for 16+c bit */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up carry.. */
+	x = (x & 0xffff) + (x >> 16);
+	return x;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented.
+ */
+unsigned short int csum_tcpudp_magic(unsigned long saddr,
+				   unsigned long daddr,
+				   unsigned short len,
+				   unsigned short proto,
+				   unsigned int sum)
+{
+	return ~from64to16(saddr + daddr + sum +
+		((unsigned long) ntohs(len) << 16) +
+		((unsigned long) proto << 8));
+}
+
+unsigned int csum_tcpudp_nofold(unsigned long saddr,
+				   unsigned long daddr,
+				   unsigned short len,
+				   unsigned short proto,
+				   unsigned int sum)
+{
+	unsigned long result;
+
+	result = (saddr + daddr + sum +
+		  ((unsigned long) ntohs(len) << 16) +
+		  ((unsigned long) proto << 8));
+
+	/* Fold down to 32-bits so we don't loose in the typedef-less 
+	   network stack.  */
+	/* 64 to 33 */
+	result = (result & 0xffffffff) + (result >> 32);
+	/* 33 to 32 */
+	result = (result & 0xffffffff) + (result >> 32);
+	return result;
+}
+
+/*
+ * Do a 64-bit checksum on an arbitrary memory area..
+ *
+ * This isn't a great routine, but it's not _horrible_ either. The
+ * inner loop could be unrolled a bit further, and there are better
+ * ways to do the carry, but this is reasonable.
+ */
+static inline unsigned long do_csum(const unsigned char * buff, int len)
+{
+	int odd, count;
+	unsigned long result = 0;
+
+	if (len <= 0)
+		goto out;
+	odd = 1 & (unsigned long) buff;
+	if (odd) {
+		result = *buff << 8;
+		len--;
+		buff++;
+	}
+	count = len >> 1;		/* nr of 16-bit words.. */
+	if (count) {
+		if (2 & (unsigned long) buff) {
+			result += *(unsigned short *) buff;
+			count--;
+			len -= 2;
+			buff += 2;
+		}
+		count >>= 1;		/* nr of 32-bit words.. */
+		if (count) {
+			if (4 & (unsigned long) buff) {
+				result += *(unsigned int *) buff;
+				count--;
+				len -= 4;
+				buff += 4;
+			}
+			count >>= 1;	/* nr of 64-bit words.. */
+			if (count) {
+				unsigned long carry = 0;
+				do {
+					unsigned long w = *(unsigned long *) buff;
+					count--;
+					buff += 8;
+					result += carry;
+					result += w;
+					carry = (w > result);
+				} while (count);
+				result += carry;
+				result = (result & 0xffffffff) + (result >> 32);
+			}
+			if (len & 4) {
+				result += *(unsigned int *) buff;
+				buff += 4;
+			}
+		}
+		if (len & 2) {
+			result += *(unsigned short *) buff;
+			buff += 2;
+		}
+	}
+	if (len & 1)
+		result += *buff;
+	result = from64to16(result);
+	if (odd)
+		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
+out:
+	return result;
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl)
+{
+	return ~do_csum(iph,ihl*4);
+}
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum)
+{
+	unsigned long result = do_csum(buff, len);
+
+	/* add in old sum, and carry.. */
+	result += sum;
+	/* 32+c bits -> 32 bits */
+	result = (result & 0xffffffff) + (result >> 32);
+	return result;
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+unsigned short ip_compute_csum(unsigned char * buff, int len)
+{
+	return ~from64to16(do_csum(buff,len));
+}
diff --git a/pfinet/linux-src/include/asm-x86_64/checksum.h b/pfinet/linux-src/include/asm-x86_64/checksum.h
new file mode 100644
index 00000000..a4238ef1
--- /dev/null
+++ b/pfinet/linux-src/include/asm-x86_64/checksum.h
@@ -0,0 +1,109 @@
+#ifndef _ASM_X86_CHECKSUM_64_H
+#define _ASM_X86_CHECKSUM_64_H
+
+/*
+ * Checksums for x86-64
+ * Copyright 2002 by Andi Kleen, SuSE Labs
+ * with some code from asm-x86/checksum.h
+ */
+
+#include <asm/uaccess.h>
+#include <asm/byteorder.h>
+
+/*
+ *	Fold a partial checksum without adding pseudo headers
+ */
+
+static inline unsigned short csum_fold(unsigned int sum)
+{
+	sum = (sum & 0xffff) + (sum >> 16);
+	sum = (sum & 0xffff) + (sum >> 16);
+	return ~sum;
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+extern unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl);
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+extern unsigned short int csum_tcpudp_magic(unsigned long saddr,
+					   unsigned long daddr,
+					   unsigned short len,
+					   unsigned short proto,
+					   unsigned int sum);
+
+unsigned int csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
+				unsigned short len, unsigned short proto,
+				unsigned int sum);
+
+/**
+ * csum_partial - Compute an internet checksum.
+ * @buff: buffer to be checksummed
+ * @len: length of buffer.
+ * @sum: initial sum to be added in (32bit unfolded)
+ *
+ * Returns the 32bit unfolded internet checksum of the buffer.
+ * Before filling it in it needs to be csum_fold()'ed.
+ * buff should be aligned to a 64bit boundary if possible.
+ */
+extern unsigned int csum_partial(const void *buff, int len, unsigned int sum);
+
+#define  _HAVE_ARCH_COPY_AND_CSUM_FROM_USER 1
+#define HAVE_CSUM_COPY_USER 1
+
+static inline unsigned int 
+csum_partial_copy(const char *src, char *dst, int len,unsigned int sum)
+{
+	memcpy(dst,src,len);
+        return csum_partial(dst, len, sum);
+}
+
+/* Do not call this directly. Use the wrappers below */
+static inline unsigned int
+csum_partial_copy_generic(const void *src, const void *dst,
+			  int len, unsigned int sum,
+			  int *src_err_ptr, int *dst_err_ptr)
+{
+	return csum_partial_copy(src, dst, len, sum);
+}
+
+static __inline__
+unsigned int csum_partial_copy_to_user ( const void *src, void *dst,
+					 int len, unsigned int sum, int *err_ptr)
+{
+	return csum_partial_copy_generic ( src, dst, len, sum, NULL, err_ptr);
+}
+
+static __inline__
+unsigned int csum_partial_copy_from_user ( const void *src, void *dst,
+					   int len, unsigned int sum, int *err_ptr)
+{
+	return csum_partial_copy_generic ( src, dst, len, sum, err_ptr, NULL);
+}
+
+extern unsigned int csum_partial_copy_nocheck(const void *src, void *dst,
+					      int len, unsigned int sum);
+
+#define csum_partial_copy_nocheck(src, dst, len, sum)	\
+	csum_partial_copy((src), (dst), (len), (sum))
+
+/* Old names. To be removed. */
+#define csum_and_copy_to_user csum_partial_copy_to_user
+#define csum_and_copy_from_user csum_partial_copy_from_user
+
+/**
+ * ip_compute_csum - Compute an 16bit IP checksum.
+ * @buff: buffer address.
+ * @len: length of buffer.
+ *
+ * Returns the 16bit folded/inverted checksum of the passed buffer.
+ * Ready to fill in.
+ */
+extern unsigned int ip_compute_csum(const void *buff, int len);
+
+#endif /* _ASM_X86_CHECKSUM_64_H */